Optimization of Recovery Processes for Multiple Economic and Environmental Objectives

Of the three spheres of sustainability-economic development, environmental stewardship, and societal equity, only the first two are quantifiable based on process variables. While economic criteria such as profit before taxes, payback period, and net present worth are well established, environmental objectives are recent and there is no general consensus on aggregation methods for calculating environmental impact. Many contributing factors have been identified for environmental impacts: impact on humans, ecosystem-terrestrial and aquatic, and local/global temperatures-global warming and ozone depletion, as well as photochemical oxidation, acid rain, and eutrophication. However, reported studies have used one aggregate environmental index as the objective in process design besides one economic objective. Hence, feasibility and usefulness of process optimization for more than two economic and environmental objectives is studied. Two case studies: a VOC (volatile organic component) recovery system and a solvent recovery system are chosen and optimized for both economic and environmental objectives using the elitist nondominated sorting genetic algorithm. The contributing factors to the environmental impacts are optimized individually or grouped into a few indices where appropriate. The objectives chosen for each round of optimization is two, several, or more, with at least one economic and one environmental objective. As they are partially or totally conflicting, Pareto-optimal solutions are obtained. These elucidate the trade-offs present, and the decision maker would be better equipped in choosing the best solution. The net flow method is then used to identify the best Pareto-optimal solution, whereby the decision maker's preference has to be declared. Pareto-optimal Solutions and the best Pareto-optimal solution for the two case studies are presented and discussed. Insights gained from considering a number of environmental objectives for process optimization are highlighted.